What Hormone Therapy Does Well

Before getting into the WHI story, it's worth establishing what menopausal hormone therapy (MHT) actually does. The evidence for its primary indication is among the strongest for any intervention in women's health.

75%
reduction in hot flash frequency with estrogen therapy — more effective than any non-hormonal alternative [1]
87%
reduction in hot flash severity with estrogen therapy [2]
21–34%
reduction in fracture risk across vertebral, hip, and non-vertebral sites [3]
📊 The Evidence — What MHT Treats Well

A Cochrane meta-analysis of 24 randomized controlled trials found a 75% reduction in weekly hot flash frequency with estrogen alone or combined with progestogen, with benefit typically appearing within two weeks at standard doses.[1] No non-hormonal treatment comes close — SSRIs, SNRIs, and gabapentin reduce vasomotor symptoms by 40–65% at best.[2]

For genitourinary syndrome of menopause — vaginal dryness, pain with intercourse, urinary urgency — local vaginal estrogen is the most effective treatment available and is appropriate even for women who cannot or choose not to take systemic MHT. Sleep disturbance improves when driven by night sweats, though MHT doesn't improve sleep in women without vasomotor symptoms.[4] Bone density is meaningfully preserved, though the benefit dissipates within about five years of stopping — which means MHT isn't the preferred long-term strategy for osteoporosis prevention.[3][4]

For mood: transdermal estradiol with micronized progesterone showed prophylactic benefit for depressive symptoms during the menopause transition in one well-designed randomized trial — clinically significant depressive symptoms occurred in 17% of the hormone group vs. 32% of placebo, independent of vasomotor symptom improvement.[5] MHT is not recommended as primary treatment for major depression, but the mood connection during the transition itself is real.

The WHI: What Happened and What Went Wrong

In 2002, the Women's Health Initiative — a large randomized trial funded by the National Institutes of Health — stopped its combined estrogen-plus-progestin arm early after finding increased rates of breast cancer, heart disease, stroke, and blood clots in the treatment group. The announcement generated enormous media coverage and triggered a dramatic, rapid decline in hormone therapy prescribing. Between 2002 and 2004, MHT use in the US dropped by more than 50%.

The findings were real. But the way they were communicated and interpreted missed several critical nuances — and the overcorrection that followed led to a decade in which millions of women with significant vasomotor symptoms went untreated, often needlessly.

The problem: Who was in the trial

Average age 63 — not the typical MHT patient

The WHI enrolled women with an average age of 63, many of whom had been postmenopausal for over a decade. The typical woman starting MHT for symptom relief is in her early 50s, recently menopausal. Applying findings from a much older population to that group was a fundamental misapplication of the data.

The problem: Relative vs. absolute risk

Headlines reported relative risk; absolute risk was small

The increased breast cancer risk was reported as a meaningful relative increase. But the absolute numbers — 9 extra breast cancer cases per 10,000 women per year — were far less alarming. Most women and many providers never heard the absolute figure.

The problem: Only one formulation studied

Oral conjugated equine estrogen + synthetic progestin

The WHI used oral conjugated equine estrogens and medroxyprogesterone acetate — a synthetic progestin. These are not the only or even the preferred formulations today. Transdermal estradiol and micronized progesterone have different metabolic and risk profiles, but they weren't studied in WHI and their results were initially lumped together with it.

The problem: Estrogen-only arm told a different story

Estrogen alone actually reduced breast cancer risk

The WHI also included an estrogen-only arm for women who had had hysterectomies. That arm was not stopped early for harm — it was stopped for lack of cardiovascular benefit in an older population. And unlike the combined arm, estrogen-only therapy actually reduced breast cancer incidence. This finding received a fraction of the attention the combined arm did.

The consequence

A decade of undertreatment

MHT prescribing plummeted and didn't recover for years. Women with severe vasomotor symptoms were counseled to avoid hormones. Some took that counsel unnecessarily. The harm of undertreating significant symptoms — disrupted sleep, quality of life, sexual health, and in surgical menopause, long-term cardiovascular and cognitive consequences — went largely unquantified.

The Timing Hypothesis: When You Start Matters

One of the most important insights to emerge from the reanalysis of WHI data and subsequent research is that the risks and benefits of MHT are not fixed — they depend heavily on when a woman starts relative to her menopause.

Women who begin MHT within 10 years of menopause or before age 60 appear to have a more favorable benefit-to-risk profile than women who start later. In the WHI, women aged 50–59 who received estrogen alone actually had fewer adverse events overall than women who received placebo. Women aged 70–79 had more. The cardiovascular story is particularly age-dependent: estrogen started close to menopause appears to be cardioprotective or neutral; started a decade or more after menopause, it may increase cardiovascular risk.

📊 The Evidence — Timing Matters

In the WHI, women aged 50–59 receiving estrogen alone had 19 fewer adverse events per 10,000 person-years compared to placebo — a net benefit. Women aged 70–79 on the same therapy had 51 more adverse events per 10,000 person-years — a net harm.[6] The difference reflects both the timing of initiation relative to menopause and the baseline cardiovascular risk that accumulates with age.

The supporting data across multiple studies is consistent. The Danish Osteoporosis Prevention Study (DOPS), which enrolled women with a mean age of 50, found that after 10 years of MHT the composite endpoint of heart failure, myocardial infarction, and death was reduced by 52% (HR 0.48; 95% CI 0.26–0.87) with no increase in stroke.[7] A Finnish registry study of nearly 500,000 women found significantly lower coronary heart disease mortality in women who started MHT at younger ages. Pooled WHI subgroup analysis found a 31% reduction in all-cause mortality in women aged 50–59 when both MHT arms were combined.[7]

The ELITE trial provided mechanistic support: estradiol therapy slowed progression of carotid intima-media thickness — a marker of subclinical atherosclerosis — when started within 6 years of menopause, but had no effect when started 10 or more years after menopause.[8] The biological explanation is plausible: estrogen appears to protect arterial walls when the vasculature is still relatively healthy, but may accelerate plaque instability when significant atherosclerotic disease is already established.

The USPSTF's 2022 systematic review acknowledges the timing hypothesis but notes that no randomized trial has prospectively tested timing of initiation as a primary question — the subgroup analyses are hypothesis-generating, not confirmatory.[9] The current guidelines do not recommend starting MHT specifically for cardiovascular prevention. VTE risk also remains elevated even in younger women regardless of timing, though transdermal estrogen substantially mitigates this.[7]

My Synthesis My clinical approach is straightforward: I treat symptoms. If a woman has significant vasomotor symptoms and no contraindications, there is no good reason to delay MHT while waiting for more cardiovascular data — and the timing data makes a compelling case for starting sooner rather than later. The DOPS result, the ELITE mechanistic finding, and the WHI subgroup analysis all point in the same direction. I don't promise cardiovascular benefit — that's not the indication and the guidelines are clear on that — but I do tell patients that starting early when symptoms warrant it appears to be the right call on multiple fronts. What I'm not doing is starting MHT in a 68-year-old who has been postmenopausal for 15 years and has no significant symptoms. That's a different risk profile entirely, and the evidence there is less favorable. The timing matters — which is exactly why I don't tell symptomatic patients to wait.

Formulation Matters: Not All MHT Is the Same

One of the most important practical lessons from the post-WHI era is that the specific formulation of MHT — the type of estrogen, the type of progestogen if used, and the route of delivery — influences both the benefits and the risk profile meaningfully.

Transdermal estradiol

  • Absorbed through skin; bypasses liver metabolism
  • Lower VTE risk than oral estrogen — avoids first-pass hepatic effects that activate clotting factors
  • More stable blood levels than oral formulations
  • Available as patches, gels, or sprays
  • Preferred route in women with cardiovascular risk factors or VTE history

Oral estrogen

  • Passes through liver before entering circulation
  • Higher VTE risk due to hepatic first-pass activation of clotting factors
  • Slightly higher total cholesterol but also raises HDL
  • Easier to dose; well-studied
  • Significantly less expensive than patches — an important practical consideration for many patients
  • Reasonable choice in lower-risk women without VTE history or cardiovascular risk factors

Micronized progesterone

  • Bioidentical — identical structure to body's own progesterone
  • Lower breast cancer risk signal than synthetic progestins
  • May have favorable effects on sleep and mood
  • Required in women with a uterus to protect the endometrium
  • Not needed in women without a uterus

Synthetic progestins (e.g., MPA)

  • Used in the original WHI trial
  • Higher breast cancer risk signal than micronized progesterone
  • May attenuate some cardiovascular benefits of estrogen
  • Still widely used and effective for endometrial protection
  • May be appropriate in some clinical situations
📊 The Evidence — Formulation Differences Observational data — primarily from large French and UK cohort studies — suggest that micronized progesterone and dydrogesterone carry lower breast cancer risk than synthetic progestins like medroxyprogesterone acetate, and that transdermal estrogen carries lower VTE risk than oral estrogen.[9][2] These findings are consistent and biologically plausible, but they come from observational studies rather than randomized trials — which means they could reflect prescribing patterns rather than true drug effects. The randomized trial data we have comes primarily from oral conjugated equine estrogen with synthetic progestin, which is no longer considered the preferred formulation. This is an important caveat: the risk profile of modern preferred formulations may be meaningfully better than what WHI measured, but we don't yet have definitive RCT evidence to confirm that.
My Synthesis For most healthy women going through perimenopause or early menopause who have bothersome vasomotor symptoms and no significant contraindications, my recommendation is hormone therapy — not as a last resort after non-hormonal options have failed, but as first-line treatment. The efficacy advantage is substantial (75–87% reduction in hot flash frequency and severity vs. 40–65% for the best non-hormonal alternatives), and for women in the favorable timing window the overall safety profile is good. Telling a healthy 51-year-old with severe hot flashes to try an antidepressant first, when hormone therapy is dramatically more effective and appropriate for her risk profile, is not evidence-based caution — it's overcorrection from the WHI misreading.

On formulation: I prescribe micronized progesterone over synthetic progestins for women with a uterus. The observational evidence pointing toward lower breast cancer risk is consistent, the biological rationale is sound, and the sleep and mood benefits are a real clinical plus. For estrogen, my default is transdermal for women with any cardiovascular risk factors or VTE history — but for low-risk women, oral estradiol is a completely reasonable choice and is significantly cheaper than patches, which matters for long-term adherence. The right formulation is the one the patient will actually take consistently. Women without a uterus get estrogen only — no progestogen needed and the evidence suggests adding it may do more harm than good, a point I cover in detail in Post 3.3.